Full-automatic vacuum negative pressure self-suction device

Abstract

The application discloses a kind of full-automatic vacuum negative pressure self-suction device, belong to the technical field of water pump, including vacuum pump body, the water inlet end of the vacuum pump body is equipped with pressure sensor one for detecting water inlet pressure, the water outlet end of the vacuum pump body is equipped with pressure sensor two for detecting water outlet pressure, controller is installed on the outer side of the vacuum pump body, the controller is configured to perform the following module functions: acquisition module is used to obtain the pressure signal of water inlet end and water outlet end, and carries out noise suppression and stability processing, to generate stable water inlet pressure value, water outlet pressure value, and calculate real-time pressure difference, pressure difference change trend and flow correlation parameter accordingly, and then construct feature vector;Diagnosis module is used for fault identification according to feature vector. Early detection, early determination of failure can be realized, effectively avoid the serious loss caused by fault expansion, improve the stability and reliability of device operation.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of water suction pumps, in particular to a full-automatic vacuum negative pressure self-suction device. BACKGROUND

[0002] In outdoor drainage, a vacuum pump station is generally used, which includes important equipment such as a vacuum tank, a negative pressure vacuum pump, and a sewage pump. These devices work together to ensure the smooth operation of outdoor drainage and meet the drainage needs in different scenarios. In actual operation, the performance stability and efficiency of the negative pressure vacuum pump, as one of the core components of the vacuum pump station, directly affect the operation effect of the entire drainage system. However, the negative pressure vacuum pump may face various problems during long-term operation, affecting its normal working state and the overall efficiency of the drainage system.

[0003] On the one hand, the stability of the water output of the negative pressure vacuum pump is difficult to accurately control. In actual working scenarios, although the water output per unit time is set, due to various factors such as system pressure fluctuations and changes in pipeline resistance, the actual water output may deviate. A decrease in water output will lead to a decrease in drainage efficiency, making it difficult to effectively remove accumulated water in time, which may cause a series of problems, such as road waterlogging in urban drainage systems, affecting the normal operation of the city; in industrial drainage scenarios, it may affect the normal operation of production equipment, causing economic losses, and it is difficult to adjust the water pump operation state in time to ensure that the water output is stable at the set value.

[0004] On the other hand, the problem of internal blockage of the negative pressure vacuum pump occurs from time to time. When the negative pressure vacuum pump pumps sewage and other media, impurities and particulate matter in the sewage can easily accumulate inside the negative pressure vacuum pump, causing blockage. Once the negative pressure vacuum pump is blocked, it will seriously affect the pumping capacity of the water pump, leading to poor drainage, and increase the operating load of the negative pressure vacuum pump, increasing its energy consumption, and even damaging the pump body equipment, shortening its service life. The existing technical means cannot accurately detect blockage in the early stage of pump body blockage. The blockage problem is often discovered when it has a significant impact on the drainage system, which not only increases maintenance costs and difficulty, but also greatly interferes with the normal operation of the drainage system. SUMMARY

[0005] To solve the problems in the prior art, the present application aims to provide a full-automatic vacuum negative pressure self-suction device that can detect faults early and accurately, effectively avoid serious losses caused by fault expansion, and improve the stability and reliability of the device operation.

[0006] To solve the above problems, the present application adopts the following technical solutions.

[0007] A full-automatic vacuum negative pressure self-suction device, comprising a vacuum pump body, a pressure sensor one for detecting the water inlet pressure is installed at the water inlet end of the vacuum pump body, a pressure sensor two for detecting the water outlet pressure is installed at the water outlet end of the vacuum pump body, a controller is installed on the outer side of the vacuum pump body, and the controller is configured to perform the following module functions:

[0008] The acquisition module is used to acquire the pressure signals of the water inlet end and the water outlet end, and to perform noise suppression and stability processing to generate stable water inlet pressure and water outlet pressure values, and to calculate real-time pressure difference, pressure difference trend and flow correlation parameters, and to construct a feature vector;

[0009] The diagnosis module is used to identify faults according to the feature vector, and the fault identification includes: generating a blockage identification based on the correlation between the abnormal increase of the water inlet pressure and the abnormal decrease of the water outlet pressure, generating a leakage identification based on the coupling of the pressure difference mutation and the deviation of the flow correlation parameter, and generating a mechanical abnormality identification through dynamic feature pattern recognition, and outputting the fault classification result and the health quantitative evaluation value;

[0010] The dynamic control module is used to trigger adaptive rules according to the fault classification result and the health quantitative evaluation value when abnormal conditions are identified, generate a preliminary power control amount, and calculate a power boundary based on the health quantitative evaluation value;

[0011] The decision module is used to minimize the comprehensive cost function of pump operation as the optimization objective, and to optimize the preliminary power control amount under the power boundary constraint through an optimization algorithm to generate a power decision instruction;

[0012] The evaluation module is used to input the power decision instruction into a simulation model for dynamic response prediction, and to modify the power decision instruction when the predicted parameters exceed the safety boundary, and to generate a remaining life assessment based on a pump body degradation model;

[0013] The iteration module is used to output the instruction to the execution unit, and to trigger the control strategy reset when the flow control deviation exceeds the adaptive range, and to store the running data to iteratively update the models in the diagnosis module and the evaluation module.

[0014] Further, the acquisition module constructs the feature vector by the following way:

[0015] The original pressure signals collected by the pressure sensor one and the pressure sensor two are sequentially filtered, smoothed and stability processed to obtain stable water inlet pressure and water outlet pressure values;

[0016] The real-time pressure difference between the stable water inlet pressure value and the water outlet pressure value and its trend are calculated;

[0017] Based on stable inlet and outlet water pressure values ​​and a preset physical model, calculate flow rate correlation parameters;

[0018] Real-time differential pressure, differential pressure change trends, and flow-related parameters are integrated into a feature vector.

[0019] Furthermore, the diagnostic module generates blockage and leakage indicators, including:

[0020] The stable inlet water pressure value is compared with the preset normal range to determine whether it has increased abnormally, and the stable outlet water pressure value is compared with the preset normal range to determine whether it has decreased abnormally. If the correlation between the two meets the preset blockage judgment conditions, a blockage indicator is generated.

[0021] Determine whether there is a sudden change in real-time differential pressure and whether the flow-related parameters deviate from the normal range. If the coupling effect of the two meets the preset leakage judgment conditions, a leakage indicator is generated.

[0022] Furthermore, the diagnostic module is also used for:

[0023] Dynamic features related to the mechanical operating state are extracted from the feature vector and matched with a preset normal mode feature library. If the matching degree is lower than a preset threshold, a mechanical anomaly identifier is generated.

[0024] The fault classification results are generated based on the blockage, leakage and mechanical abnormality indicators, and the health quantitative assessment value is calculated in combination with the health quantitative assessment standard.

[0025] Furthermore, the dynamic control module calculates the power boundary, including:

[0026] The baseline power limit parameters are determined based on the quantitative health assessment values.

[0027] The basic power limit parameters are corrected based on the fault classification results and verified by combining the changing trend of the health quantitative assessment value to determine the final power boundary.

[0028] Furthermore, the decision module generates power decision instructions including:

[0029] Within the feasible range defined by the power boundary, multiple candidate power adjustment schemes are generated through optimization algorithms;

[0030] Calculate the comprehensive cost value for each candidate power adjustment scheme, and select the scheme with the minimum comprehensive cost value as the final power adjustment scheme;

[0031] The final power adjustment scheme is converted into a power decision command.

[0032] Furthermore, the evaluation module performs dynamic response prediction and instruction correction, including:

[0033] Input the power decision command and current operating status data into the system simulation model to obtain the predicted values ​​of key operating parameters;

[0034] The predicted value is compared with the safety boundary. If the value exceeds the boundary, the power decision command is corrected according to the preset rules.

[0035] Furthermore, the assessment module generates a remaining lifetime assessment, including:

[0036] The power decision command, historical operating data, and current health quantitative assessment value are input into the pump body degradation model to predict the performance degradation trend;

[0037] The remaining life assessment value is calculated based on the predicted performance degradation trend and the preset failure judgment criteria.

[0038] Furthermore, the iteration module triggers a control strategy reset, including:

[0039] After outputting power commands to the execution unit, the deviation between the actual flow rate and the target flow rate is monitored;

[0040] If the deviation continues to exceed the preset adaptive range, the control strategy will be reset, and the current strategy will be replaced with the initial control strategy.

[0041] Furthermore, the iteration module also includes:

[0042] Store power commands, flow data, pressure data, and quantitative health assessment values ​​during operation;

[0043] The stored data is used to update the fault identification rules in the diagnostic module and the pump degradation model in the evaluation module.

[0044] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0045] (1) This solution uses the acquisition module to accurately process the pressure signal and construct the feature vector. Combined with the multi-dimensional fault identification logic of the diagnostic module, it can simultaneously and accurately identify three types of faults: blockage, leakage and mechanical abnormality, and generate a health quantitative assessment value. This enables early detection and early judgment of faults, effectively avoids serious losses caused by the expansion of faults, improves the stability and reliability of the device operation, and reduces production losses caused by sudden shutdowns.

[0046] (2) This scheme relies on the adaptive power adjustment and power boundary constraints of the dynamic control module, and is combined with the optimization algorithm of the decision module with the goal of minimizing the overall cost. Under the premise of ensuring the safe operation of the device, it can realize intelligent dynamic power adjustment, balance energy consumption cost, fault loss and maintenance needs, significantly reduce the overall operating cost, and improve the economy and adaptability of the device operation.

[0047] (3) The evaluation module of this scheme uses the system simulation model to verify the safety of the power command and correct the over-limit, thereby avoiding safety risks from the source. At the same time, it uses the pump body degradation model to accurately predict the remaining life, providing a scientific basis for the formulation of maintenance plans, realizing forward-looking maintenance, reducing the frequency of unplanned downtime, extending the overall service life of the device, and reducing maintenance costs and resource waste.

[0048] (4) The closed-loop self-learning mechanism of the iterative module of this scheme continuously stores the running data and iteratively updates the diagnostic rules and degradation model, so that the fault identification accuracy, life prediction accuracy and control strategy adaptability of the device gradually improve with the running time. It has self-optimization and self-adaptation capabilities, can flexibly cope with complex working condition changes, and does not require frequent manual adjustment, further improving the intelligence level and long-term operating benefits of the device. Attached Figure Description

[0049] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0050] Fig. 1 This is an overall structural view of a fully automatic vacuum negative pressure self-priming device according to the present invention;

[0051] Fig. 2 This is a control flowchart of a fully automatic vacuum negative pressure self-priming device according to the present invention.

[0052] Explanation of the labels in the diagram:

[0053] 1. Vacuum pump body; 2. Pressure sensor one; 3. Pressure sensor two; 4. Controller. Detailed Implementation

[0054] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0055] Please see Figs. 1-2A fully automatic vacuum negative pressure self-priming device includes a vacuum pump body 1, characterized in that: a pressure sensor 2 for detecting inlet water pressure is installed at the water inlet end of the vacuum pump body 1, a pressure sensor 3 for detecting outlet water pressure is installed at the water outlet end of the vacuum pump body 1, and a controller 4 is installed on the outer surface of the vacuum pump body 1, the controller 4 being configured to perform the functions of the following modules:

[0056] The acquisition module is used to acquire pressure signals at the inlet and outlet ends, and perform noise suppression and stability processing to generate stable inlet and outlet pressure values. Based on these values, it calculates real-time pressure difference, pressure difference change trend and flow-related parameters, and then constructs feature vectors.

[0057] The diagnostic module is used to identify faults based on feature vectors. Fault identification includes: generating blockage indicators based on the correlation between abnormal increase in inlet water pressure and abnormal decrease in outlet water pressure; generating leakage indicators based on the coupling between pressure difference change and deviation of flow-related parameters; and generating mechanical abnormality indicators through dynamic feature pattern recognition. The module outputs fault classification results and health quantification assessment values.

[0058] The dynamic control module is used to trigger adaptive rules based on fault classification results and health quantification assessment values ​​when abnormal operating conditions are detected, generate preliminary power control quantities, and calculate power boundaries based on health quantification assessment values.

[0059] The decision module is used to optimize the initial power control quantity through an optimization algorithm under power boundary constraints, with the optimization objective of minimizing the comprehensive cost function of pump operation, and to generate power decision instructions.

[0060] The evaluation module is used to input power decision commands into the simulation model for dynamic response prediction, correct the power decision commands when the predicted parameters exceed the safety boundary, and generate a remaining life assessment based on the pump body degradation model.

[0061] The iteration module is used to output instructions to the execution unit and trigger a control strategy reset when the flow control deviation exceeds the adaptive range. At the same time, it stores running data to iteratively update the models in the diagnostic and evaluation modules.

[0062] In addition, the acquisition module constructs feature vectors in the following way:

[0063] The raw pressure signals collected by pressure sensor 12 and pressure sensor 23 are sequentially filtered, smoothed and stabilized to obtain stable inlet and outlet water pressure values.

[0064] Calculate the real-time pressure difference and its trend between stable inlet and outlet water pressure values;

[0065] Based on stable inlet and outlet water pressure values ​​and a preset physical model, calculate flow rate correlation parameters;

[0066] Real-time differential pressure, differential pressure change trends, and flow-related parameters are integrated into a feature vector.

[0067] In this embodiment, pressure sensor 2 and pressure sensor 3 are both sensors for measuring water pressure. During operation, pressure sensor 2 at the inlet and pressure sensor 3 at the outlet synchronously collect raw pressure signals. These signals inevitably contain irrelevant components such as environmental electromagnetic interference and sensor noise. Therefore, the raw signals need to be systematically processed. First, high-frequency interference signals are removed by filtering, then random fluctuations in the signals are reduced by smoothing, and finally, stability verification is performed to ensure that the data is not affected by instantaneous changes. Finally, stable inlet and outlet pressure values ​​that can truly reflect the actual operating state of the device are obtained. Based on the stable pressure data, the real-time pressure difference is first calculated. This calculation is based on the basic physical concept of pressure difference between two points in fluid flow, and the formula is abstracted through direct subtraction. ,in This represents the stable inlet water pressure value. This represents the stable outlet water pressure value. This refers to the real-time pressure difference between the inlet and outlet. The real-time pressure difference directly reflects the pressure difference between the two ends and is a fundamental parameter for judging the fluid flow resistance.

[0068] To capture the dynamic characteristics of differential pressure changes, it is necessary to further calculate the trend of differential pressure changes. Considering that discrete sampling methods are often used to acquire data in engineering practice, making direct continuous-time differentiation impossible, this paper uses the approximate calculation method of instantaneous rate of change in calculus. The ratio of the differential pressure difference between two adjacent sampling times to the time interval is used to abstract a quantitative index of the differential pressure change trend in discrete time. The corresponding formula is: ,in This represents the real-time pressure difference at the current moment. This represents the real-time pressure difference at the previous sampling time. For a fixed sampling time interval, This is a quantitative value of the pressure difference change trend, which can effectively reflect the rise, fall or stabilization of the pressure difference over time.

[0069] The calculation of flow-related parameters relies on a pre-defined physical model, which is based on pipe flow theory in fluid mechanics. For example, for a specific specification of conveying pipe and a fluid with known characteristics, there is a clear linear relationship between flow rate and pressure difference under laminar flow conditions. By integrating fixed parameters such as pipe inner diameter, length, and fluid viscosity into a pre-defined coefficient k, which is jointly determined by pipe structural parameters and fluid physical properties, complex pipe flow formulas can be transformed. Simplified formula for calculating flow-related parameters for practical engineering applications .in As a flow-related parameter, this simplified formula ensures both the effective characterization of flow-related properties and meets the efficiency requirements of real-time calculation by the device.

[0070] Finally, the acquisition module will calculate the real-time pressure difference. Pressure difference change trend and traffic-related parameters By integrating these elements, a feature vector with complete dimensions and complementary information can be formed. This feature vector comprehensively covers the static values, dynamic changes, and flow-related characteristics of the device's pressure status.

[0071] In some embodiments of the present invention, the diagnostic module generates blockage and leakage indicators including:

[0072] The stable inlet water pressure value is compared with the preset normal range to determine whether it has increased abnormally, and the stable outlet water pressure value is compared with the preset normal range to determine whether it has decreased abnormally. If the correlation between the two meets the preset blockage judgment conditions, a blockage indicator is generated.

[0073] Determine whether the real-time differential pressure changes abruptly and whether the flow-related parameters deviate from the normal range. If the coupling effect of the two meets the preset leakage judgment conditions, a leakage indicator is generated.

[0074] Dynamic features related to the mechanical operating state are extracted from the feature vector and matched with a preset normal mode feature library. If the matching degree is lower than a preset threshold, a mechanical anomaly identifier is generated.

[0075] The fault classification results are generated based on the blockage, leakage and mechanical abnormality indicators, and the health quantitative assessment value is calculated in combination with the health quantitative assessment standard.

[0076] In this embodiment, the generation of the blockage identifier is based on the physical characteristic of obstructed fluid flow. When the internal pipeline of the device is blocked, the obstruction of fluid flow will cause pressure to accumulate at the inlet, while the pressure at the outlet will drop due to insufficient fluid supply. This correlation between the increase in inlet pressure and the decrease in outlet pressure is the logic for judging the blockage fault. The diagnostic module first retrieves the preset normal range of inlet pressure and normal range of outlet pressure, and compares the stable inlet pressure value and stable outlet pressure value processed by the acquisition module with the corresponding normal range to determine whether they meet the single conditions of abnormal increase and abnormal decrease, respectively. Then, the correlation between the two is verified by the preset blockage judgment rules. For example, when the inlet pressure exceeds the upper limit of normal and the outlet pressure is lower than the lower limit of normal, and the duration of this double abnormal state reaches the preset threshold, it meets the blockage judgment condition. At this time, the diagnostic module generates a blockage identifier to mark that the device has a blockage fault.

[0077] The generation of the leak indicator relies on the coupling effect of pressure differential change and flow-related parameter deviation. Its physical logic stems from the dual impact of leakage on fluid pressure and flow rate: when a leak occurs in the device's pipeline or connection, it will cause the internal pressure balance of the system to be suddenly broken, which manifests as an instantaneous change in real-time pressure differential. At the same time, the leak will cause actual flow loss, causing the flow-related parameters calculated based on pressure data to deviate from the normal operating range. The diagnostic module first judges whether the pressure differential exceeds the preset change threshold in a short period of time through real-time monitoring, and at the same time checks whether the flow-related parameters deviate from their normal range. Then, it analyzes the coupling effect of the two according to the preset leak judgment conditions. For example, if the direction of the pressure differential change is consistent with the trend of the flow-related parameter deviation, and the degree of abnormality of both reaches the preset ratio, the coupling judgment requirements are met, and the diagnostic module then generates a leak indicator to clarify that there is a leak-related fault in the device.

[0078] The generation of mechanical anomaly indicators is based on the principle of dynamic feature pattern matching. The parameters contained in the feature vector, such as real-time differential pressure and differential pressure change trends, indirectly reflect changes in mechanical operating conditions, such as pump bearing wear and rotor imbalance. These changes form specific dynamic feature combinations. The diagnostic module extracts dynamic features directly related to mechanical operation from the feature vector, such as the fluctuation frequency and peak interval of differential pressure change trends. Then, it performs similarity matching between these extracted dynamic features and a preset normal mode feature library. The normal mode feature library is constructed from long-term operating data of the device under fault-free normal operating conditions and contains feature patterns corresponding to various normal mechanical states. During the matching process, the similarity between the extracted features and various normal modes in the feature library is calculated. If the similarity is lower than the preset matching threshold, it indicates that there is a significant deviation between the current mechanical operating state and the normal state. The diagnostic module then generates a mechanical anomaly indicator to indicate that there is a mechanical fault in the device.

[0079] The generation of fault classification results involves the integrated analysis of three types of fault indicators. The diagnostic module identifies the presence of blockage, leakage, and mechanical abnormality indicators to determine whether the device is currently in a single fault state, a compound fault state, or a fault-free state, thus outputting a clear fault classification result. The calculation of the health quantitative assessment value is based on the fault classification result, combined with preset health quantitative assessment standards, and achieves quantitative representation through a weighted summation method. The formula is as follows: The derivation logic of this formula stems from the negative correlation between health status and fault severity. When the device is fault-free, the health value is 1; the more severe the fault, the lower the health value. This is achieved by assigning different weights to the three types of faults. , , This reflects the degree of impact of different faults on the overall health of the device, with values ​​ranging from 0 to 1 and summing to 1; and is combined with the severity coefficient corresponding to each fault. , , The degree of fault abnormality is quantified, with a value range of 0-1, where 0 represents no fault and 1 represents the most severe fault. After weighted calculation, the health quantitative assessment value H is obtained, with a value range of 0-1. The closer it is to 1, the better the health status of the device. This quantitative value intuitively reflects the current health level of the device.

[0080] In some embodiments of the present invention, the dynamic control module calculates the power boundary by:

[0081] The baseline power limit parameters are determined based on the quantitative health assessment values.

[0082] The basic power limit parameters are corrected based on the fault classification results and verified by combining the changing trend of the health quantitative assessment value to determine the final power boundary.

[0083] In this embodiment, the generation of the initial power control quantity relies on an adaptive rule system. This system is built based on the differentiated power adjustment requirements of different fault types, and combines the device's tolerance capacity reflected by the health quantification assessment value. This ensures that the power adjustment can both alleviate the impact of the fault and not exacerbate equipment wear. For example, when the diagnostic module outputs a blockage indicator, the adaptive rule will tend to moderately reduce the power to reduce fluid impact and avoid wear on the blocked part or aggravation of the blockage. When a leakage indicator is output, the rule will judge the device's remaining tolerance capacity based on the health quantification assessment value. If the health status is good, a small increase in power is allowed to compensate for the flow loss caused by the leakage. If the health status is poor, the base power is maintained to avoid additional load. After the diagnostic module inputs the fault classification result and the health quantification assessment value into the dynamic control module, the module will automatically match the corresponding adaptive rule and calculate the initial power control quantity based on the current operating parameters, such as real-time flow and current power. The initial power control quantity is an initial adjustment direction and magnitude reference based on the fault mitigation requirements.

[0084] Calculating the power boundary is a crucial step in the dynamic control module. It ensures that power adjustments remain within the safe operating range of the device through quantitative constraints. First, the basic power limit parameters must be determined based on the health quantitative assessment value. This value directly reflects the device's current mechanical condition and tolerance. A higher health value indicates a wider power range the device can withstand, while a lower value necessitates tightening power limits to prevent further malfunctions. This positive correlation forms the calculation logic for the basic power limit parameters, from which the formula is abstracted. The formula is derived from the assumption of a linear correlation between health status and power tolerance limit. Extensive experimental data has verified that for every unit decrease in the quantitative health assessment value, the device's safe power limit decreases proportionally. Therefore, the maximum rated safe power of the device is used as the basis for the formula. Using this as a baseline, multiplying it by the health quantitative assessment value H, where H ranges from 0 to 1, yields the baseline power limitation parameter for the current health state. ,in The maximum safe operating power determined during the device design phase is the critical power value that ensures the equipment operates without failure.

[0085] Once the basic power limit parameters are determined, they need to be corrected based on the fault classification results. Different fault types have different safety requirements for power operation, and the correction process needs to be achieved through preset fault correction coefficients. For example, blockage faults will lead to increased fluid resistance, and excessive power can easily cause pump overload; therefore, a corresponding correction coefficient is required. A value less than 1 is used to lower the base power limit parameter; the correction factor corresponding to the mechanical anomaly flag. The coefficient is set according to the severity of the anomaly; the more severe the anomaly, the smaller the coefficient, in order to strictly limit power and avoid accelerated mechanical wear. The correction coefficient for leakage faults... The value may be greater than or less than 1 depending on the location and extent of the leak, balancing flow compensation and safe operation requirements. The corrected power limiting parameter calculation formula is as follows: ,in The correction factor corresponds to the fault type, with a value ranging from 0 to 1.2. After correction, it still needs to be verified in conjunction with the changing trend of the health quantitative assessment value. If the health quantitative assessment value shows a continuous downward trend, it indicates that the device condition is deteriorating, and the power limit needs to be further tightened. Then multiply by the trend verification coefficient, which ranges from 0.8 to 1.0; if the health value tends to stabilize, keep the corrected parameters unchanged. The final result is determined through this series of steps. This is the final power boundary, which defines the maximum allowable upper limit and minimum allowable lower limit for subsequent power adjustments, ensuring that the power commands generated by the decision module are always within a safe range.

[0086] In some embodiments of the present invention, the decision module generates power decision instructions including:

[0087] Within the feasible range defined by the power boundary, multiple candidate power adjustment schemes are generated through optimization algorithms;

[0088] Calculate the comprehensive cost value for each candidate power adjustment scheme, and select the scheme with the minimum comprehensive cost value as the final power adjustment scheme;

[0089] The final power adjustment scheme is converted into a power decision command.

[0090] In this embodiment, the power boundary provides a clear feasible range for the decision-making module. The feasible range is determined by the dynamic control module based on the device's health status and fault type, clearly defining the maximum and minimum allowable values ​​for power adjustment. This ensures that all candidate schemes are within a safe operating range. The decision-making module first calls the power boundary parameters output by the dynamic control module as constraints for the optimization algorithm. Then, it starts a preset optimization algorithm, such as particle swarm optimization or genetic algorithm, which are commonly used optimization methods in engineering. Multiple candidate power adjustment schemes are generated within the range defined by the power boundary. These candidate schemes are reasonably distributed around the initial power control value, covering both the adjustment range around the initial control value and the power values ​​at different gradients within the range, ensuring the comprehensiveness and diversity of the schemes.

[0091] The construction of the comprehensive cost function is based on the logic of minimizing operating costs. The comprehensive cost of pump operation encompasses multiple dimensions, including energy consumption costs, failure loss costs, and expected maintenance costs. Each dimension has a different degree of impact on the overall operating efficiency, which needs to be quantitatively represented through weighted summation. Based on the correlation between cost, power, and health status in engineering practice, the formula for the comprehensive cost function is abstracted. The derivation of the formula stems from the breakdown of the pump's total lifecycle cost. Energy consumption cost is positively correlated with power, failure loss cost is positively correlated with the degree to which power deviates from the optimal range, and expected maintenance cost is negatively correlated with health status and power load. This is achieved by assigning weights to these three types of costs. , , The values ​​of all three range from 0 to 1, and their sum is 1, reflecting the relative importance of each cost; energy consumption costs Failure loss cost Maintenance expected costs This is integrated into a single comprehensive cost C, realizing the transformation from multi-objective optimization to single-objective optimization; among which... The cost of electrical energy consumed by the pump unit per unit time is approximately proportional to the square of the power. The quantified cost corresponding to the increased failure or lifespan loss caused by power adjustment; the worse the health condition, the more sensitive this cost is to power changes. To predict future maintenance costs based on current power load, both excessively high and low power loads may increase maintenance frequency and costs.

[0092] The decision-making module calculates the comprehensive cost value for each candidate power adjustment scheme. During the calculation, it needs to combine the current operating parameters, such as real-time flow rate, differential pressure status, and health quantitative assessment value, and substitute them into the comprehensive cost function to obtain the specific C value of each scheme. Then, by comparing the comprehensive cost value of all candidate schemes, the scheme with the smallest C value is selected as the final power adjustment scheme. The final power adjustment scheme satisfies the safety constraints of the power boundary and can minimize the comprehensive cost of pump operation, achieving the optimal balance between energy consumption, failure risk, and maintenance requirements.

[0093] The final power adjustment scheme needs to be converted into a power decision command that the execution unit can recognize. The decision module converts the power value corresponding to the final power adjustment scheme into a standardized control signal, such as a voltage signal, frequency signal, or digital command, according to the control protocol of the device's execution unit. This ensures that the command format matches the input requirements of the execution unit, such as the frequency converter or power controller. The converted power decision command not only clarifies the target power level but also includes auxiliary control parameters such as adjustment rate and stabilization time, avoiding the impact of sudden power changes on the device and ensuring the stability and reliability of the adjustment process.

[0094] In some embodiments of the present invention, the evaluation module performs dynamic response prediction and instruction correction as follows:

[0095] Input the power decision command and current operating status data into the system simulation model to obtain the predicted values ​​of key operating parameters;

[0096] The predicted value is compared with the safety boundary. If the value exceeds the boundary, the power decision command is corrected according to the preset rules.

[0097] The power decision command, historical operating data, and current health quantitative assessment value are input into the pump body degradation model to predict the performance degradation trend;

[0098] The remaining life assessment value is calculated based on the predicted performance degradation trend and the preset failure judgment criteria.

[0099] In this embodiment, the realization of dynamic response prediction relies on a pre-built simulation model. The simulation model is based on fluid mechanics, mechanical dynamics, and control theory, and integrates information such as the device's structural parameters, fluid characteristics, and control logic. It can accurately simulate the changes in the device's operating state under different power inputs. The evaluation module first receives the power decision command output by the decision module, and at the same time collects the device's current operating state data, including key parameters such as real-time differential pressure, flow-related parameters, pump body temperature, and vibration amplitude. Both types of data are input into the simulation model. The model simulates the dynamic response process after the power command is executed through numerical calculations and outputs the predicted values ​​of key operating parameters for a period of time in the future. These predicted values ​​cover indicators directly related to safe operation, such as the peak value of differential pressure change, the upper limit of pump body operating temperature, and the vibration frequency range, and fully present the expected operating state after the power command is implemented.

[0100] The safety boundary is the critical threshold for ensuring the fault-free operation of the device. It is determined during the design phase based on material tolerance limits, structural strength, and operational reliability requirements. It covers the maximum and minimum allowable values ​​of key parameters such as differential pressure, temperature, vibration, and current. The evaluation module compares the predicted values ​​of each key parameter output by the simulation model with the corresponding safety boundary one by one to determine whether there is any out-of-bounds situation. If all predicted values ​​are within the safety boundary, it means that the power decision command is safe and no correction is required. If any parameter prediction value exceeds the safety boundary, the command correction process is initiated. The correction process follows preset rules. The rule design combines the fault classification results and the degree of out-of-bounds situation. For example, when there is a slight out-of-bounds situation, the power value is linearly fine-tuned according to the out-of-bounds ratio. When there is a severe out-of-bounds situation, the power adjustment direction is adjusted based on the fault type. At the same time, the correction range is redefined with reference to the power boundary of the dynamic control module to ensure that the corrected power decision command not only meets the operational requirements but also brings all key parameter prediction values ​​back to the safety boundary, thus avoiding safety risks from the source.

[0101] The remaining life assessment is based on the pump body degradation model. This model is constructed based on the material aging patterns, mechanical wear mechanisms, and historical fault data of the pump body. It can quantify the impact of factors such as power load, operating time, and environmental conditions on the performance degradation of the device. The assessment module takes power decision commands, historical operating data, and current health quantification values ​​as input parameters. Among them, the power decision commands reflect the load intensity of future operation and directly affect the degradation rate; historical operating data includes cumulative operating time, past fault records, and records of each power adjustment, reflecting the cumulative loss status of the device; and the current health quantification value provides an initial benchmark for predicting the degradation trend. By integrating the three types of input data, the pump body degradation model simulates the process of device performance decay over time and generates a performance degradation trend curve, such as the trajectory of the health quantification value decreasing with operating time, clearly showing the changing pattern of the device's health status over a future period of time.

[0102] The remaining lifetime assessment value is calculated based on the predicted performance degradation trend and the preset failure criteria. The failure criteria clearly define the critical conditions under which the device reaches a failure state, usually using a health quantification value below a certain fixed threshold as the criterion. Based on the linear assumption of the performance degradation trend, and verified by a large amount of experimental data, the performance degradation of the device under normal maintenance and stable operation conditions approximately follows a linear law, from which the formula for calculating the remaining lifetime is abstracted. The formula is derived from the basic relationship in linear degradation: remaining lifetime = difference between current state and failure state ÷ degradation rate. The current health quantification value H and the failure determination threshold... The difference reflects the remaining performance margin of the device from its current state to its failure state, while the degradation rate v is predicted by the pump degradation model based on the input data and represents the rate of decrease in the quantitative health assessment value per unit time. The ratio of the two is the remaining life assessment value. H takes values ​​ranging from 0 to 1. The preset failure threshold is typically set to 0.1-0.2; the value of v is measured in units of 1 / hour. The unit is hours. This value intuitively reflects the estimated operating time of the device from the current moment until it reaches a failure state, providing a scientific basis for maintenance planning and spare parts replacement.

[0103] In some embodiments of the present invention, the iteration module triggering control strategy reset includes:

[0104] After outputting power commands to the execution unit, the deviation between the actual flow rate and the target flow rate is monitored;

[0105] If the deviation continues to exceed the preset adaptive range, the control strategy will be reset and the current strategy will be replaced with the initial control strategy.

[0106] Store power commands, flow data, pressure data, and quantitative health assessment values ​​during operation;

[0107] The stored data is used to update the fault identification rules in the diagnostic module and the pump degradation model in the evaluation module.

[0108] In this embodiment, the iteration module first receives the final power decision command output by the evaluation module, adapts the format according to the signal interface specification of the execution unit, and ensures that the command can be accurately recognized and executed by execution units such as frequency converters and power controllers. After the command is output, the iteration module starts the flow deviation monitoring mechanism, collects the actual flow data during the device operation in real time, and retrieves the preset target flow value. The degree of deviation between the two is obtained through quantitative calculation. The flow deviation is calculated based on the difference between the actual flow and the target flow, and the formula is abstracted to obtain the result. The formula uses the actual flow rate. With target traffic The absolute value of the difference directly reflects the degree of deviation in flow control, avoiding misjudgments caused by the cancellation of positive and negative deviations. The actual real-time flow rate during device operation is obtained from flow detection components or based on pressure data. The target flow rate value is preset according to the working requirements of the device and is the reference parameter for flow control; This represents the flow control deviation, and its magnitude directly reflects the accuracy of flow control under the current power command.

[0109] The preset adaptive range is the fault tolerance interval for flow control, determined by factors such as the device's operational accuracy requirements and load fluctuation characteristics in engineering practice. It allows for small fluctuations in flow to adapt to complex operating conditions while preventing excessive deviations from affecting performance. The iterative module calculates the flow control deviation in real time. The system continuously compares the deviation with the adaptive range. If the deviation remains within the range, the current control strategy is effective, and the existing power command and control logic are maintained unchanged. If the deviation continuously exceeds the adaptive range and the duration reaches the preset threshold, the current control strategy can no longer meet the flow control requirements. This may be due to sudden changes in operating conditions, model parameter drift, or other reasons. At this time, the iteration module triggers the control strategy reset mechanism. The reset process replaces the current strategy with the device's initial control strategy. The initial control strategy is a basic control scheme based on the device's design parameters and standard operating conditions. It has the characteristics of strong stability and wide adaptability, and can quickly pull the device's operating state back to the controllable range to avoid the deviation from expanding and causing a chain of failures.

[0110] Throughout the entire operating cycle of the device, the iterative module continuously performs data storage tasks, storing key data covering the entire process: including output power commands, real-time collected flow data, inlet and outlet water pressure data monitored by pressure sensors, and health quantitative assessment values ​​calculated by the diagnostic module. It also includes derived data such as fault identifiers, command correction records, and remaining life assessment values. This data comprehensively records the device's operating status, control process, and health changes, forming a dataset for model iterative updates and providing real and rich sample support for subsequent optimization.

[0111] The purpose of data storage is to enable iterative optimization of the models in the diagnostic and evaluation modules. For the diagnostic module, the iterative module categorizes and organizes the stored historical operating data according to fault type and operating conditions, extracts new fault features and normal operating features, and adds them to the preset normal mode feature library. At the same time, it verifies and corrects the original fault identification rules based on the new data, such as adjusting the judgment thresholds for blockage and leakage, and optimizing the similarity standard for feature matching, so that fault identification is more in line with the actual operating characteristics of the device. For the evaluation module, the stored historical data contains the performance degradation trajectory under different power loads and fault states. The iterative module inputs this data into the pump degradation model, calibrates the key parameters such as degradation rate and influencing factors in the model through algorithms, corrects the model prediction bias, and makes the remaining life assessment value closer to the actual situation. Through this continuous data storage and model update, the fault identification accuracy, life prediction accuracy, and control strategy adaptability of the device will gradually improve with operating time, realizing a closed-loop evolution of self-learning and self-optimization.

[0112] The above are merely preferred embodiments of the present invention; however, the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and its improved concept, should be covered within the scope of protection of the present invention.

Claims

1. A fully automatic vacuum negative pressure self-priming device, comprising a vacuum pump body (1), characterized in that: The vacuum pump body (1) is equipped with a pressure sensor 1 (2) for detecting the inlet water pressure at the water inlet end, and a pressure sensor 2 (3) for detecting the outlet water pressure at the water outlet end. A controller (4) is installed on the outer surface of the vacuum pump body (1), and the controller (4) is configured to perform the functions of the following modules: The acquisition module is used to acquire pressure signals at the inlet and outlet ends, and perform noise suppression and stability processing to generate stable inlet and outlet pressure values. Based on these values, it calculates real-time pressure difference, pressure difference change trend and flow-related parameters, and then constructs feature vectors. The diagnostic module is used to identify faults based on feature vectors. Fault identification includes: generating blockage indicators based on the correlation between abnormal increase in inlet water pressure and abnormal decrease in outlet water pressure; generating leakage indicators based on the coupling between pressure difference change and deviation of flow-related parameters; and generating mechanical abnormality indicators through dynamic feature pattern recognition. The module outputs fault classification results and health quantification assessment values. The dynamic control module is used to trigger adaptive rules based on fault classification results and health quantification assessment values ​​when abnormal operating conditions are detected, generate preliminary power control quantities, and calculate power boundaries based on health quantification assessment values. The decision module is used to optimize the initial power control quantity through an optimization algorithm under power boundary constraints, with the optimization objective of minimizing the comprehensive cost function of pump operation, and to generate power decision instructions. The evaluation module is used to input power decision commands into the simulation model for dynamic response prediction, correct the power decision commands when the predicted parameters exceed the safety boundary, and generate a remaining life assessment based on the pump body degradation model. The iteration module outputs instructions to the execution unit and triggers a control strategy reset when the flow control deviation exceeds the adaptive range. It also stores runtime data to iteratively update the models in the diagnostic and evaluation modules.

2. The fully automatic vacuum negative pressure self-priming device according to claim 1, characterized in that, The acquisition module constructs feature vectors in the following way: The original pressure signals collected by pressure sensor 1 (2) and pressure sensor 2 (3) are filtered, smoothed and stabilized in sequence to obtain stable inlet and outlet pressure values. Calculate the real-time pressure difference and its trend between stable inlet and outlet water pressure values; Based on stable inlet and outlet water pressure values ​​and a preset physical model, calculate flow rate correlation parameters; Real-time differential pressure, differential pressure change trends, and flow-related parameters are integrated into a feature vector.

3. The fully automatic vacuum negative pressure self-priming device according to claim 2, characterized in that, The diagnostic module generates blockage and leakage indicators, including: The stable inlet water pressure value is compared with the preset normal range to determine whether it has increased abnormally, and the stable outlet water pressure value is compared with the preset normal range to determine whether it has decreased abnormally. If the correlation between the two meets the preset blockage judgment conditions, a blockage indicator is generated. Determine whether there is a sudden change in real-time differential pressure and whether the flow-related parameters deviate from the normal range. If the coupling effect of the two meets the preset leakage judgment conditions, a leakage indicator is generated.

4. The fully automatic vacuum negative pressure self-priming device according to claim 3, characterized in that, The diagnostic module is also used for: Dynamic features related to the mechanical operating state are extracted from the feature vector and matched with a preset normal mode feature library. If the matching degree is lower than a preset threshold, a mechanical anomaly identifier is generated. The fault classification results are generated based on the blockage, leakage and mechanical abnormality indicators, and the health quantitative assessment value is calculated in combination with the health quantitative assessment standard.

5. The fully automatic vacuum negative pressure self-priming device according to claim 4, characterized in that, The dynamic control module calculates the power boundary, including: The baseline power limit parameters are determined based on the quantitative health assessment values. The basic power limit parameters are corrected based on the fault classification results and verified by combining the changing trend of the health quantitative assessment value to determine the final power boundary.

6. The fully automatic vacuum negative pressure self-priming device according to claim 5, characterized in that, The decision module generates power decision instructions including: Within the feasible range defined by the power boundary, multiple candidate power adjustment schemes are generated through optimization algorithms; Calculate the comprehensive cost value for each candidate power adjustment scheme, and select the scheme with the minimum comprehensive cost value as the final power adjustment scheme; The final power adjustment scheme is converted into a power decision command.

7. The fully automatic vacuum negative pressure self-priming device according to claim 6, characterized in that, The evaluation module performs dynamic response prediction and instruction correction, including: Input the power decision command and current operating status data into the system simulation model to obtain the predicted values ​​of key operating parameters; The predicted value is compared with the safety boundary. If the value exceeds the boundary, the power decision command is corrected according to the preset rules.

8. The fully automatic vacuum negative pressure self-priming device according to claim 7, characterized in that, The assessment module generates a remaining lifetime assessment, including: The power decision command, historical operating data, and current health quantitative assessment value are input into the pump body degradation model to predict the performance degradation trend; The remaining life assessment value is calculated based on the predicted performance degradation trend and the preset failure judgment criteria.

9. The fully automatic vacuum negative pressure self-priming device according to claim 1, characterized in that, The iterative module triggers a control strategy reset, including: After outputting power commands to the execution unit, the deviation between the actual flow rate and the target flow rate is monitored; If the deviation continues to exceed the preset adaptive range, the control strategy will be reset, and the current strategy will be replaced with the initial control strategy.

10. The fully automatic vacuum negative pressure self-priming device according to claim 9, characterized in that, The iteration module also includes: Store power commands, flow data, pressure data, and quantitative health assessment values ​​during operation; The stored data is used to update the fault identification rules in the diagnostic module and the pump degradation model in the evaluation module.

Images